The Surgeon General of the United States has cited the prevention and treatment of the disease of obesity as a national health priority. Although rates of obesity are higher among minorities and highest among minority women, no data on the health outcome of weight control therapy in African Americans are available. We therefore propose conducting a prospective, randomized, controlled multisite (MA, NY, TX, CA) clinical trial of a culturally sensitive weight control program to produce a 5-10% weight loss in obese (BMI 3040) African American women ages 40-64. Eligible women will be randomized either to a control arm (n=200), which will receive the U.S. Dietary Guidelines to follow and will be told to attend milestone data collection visits only; to a food-based diet (FBD, n=200), which will provide 1200-/+200 kcal through a balanced diet based on the ADA exchange list; or to a meal-replacement diet (MRD, n=200), which will receive the same diet as for the FBD group but will replace exchanges for two meals with milk-based fortified meal replacement shakes. Subjects in the latter two arms will participate in the Black American Lifestyle Intervention (BALI) program, a 16-week active weight loss phase (weekly group education & exercise) followed by two years of relapse prevention (semimonthly) and weight maintenance (monthly). Milestone data will be collected at baseline and at Weeks 16, 52, and 104; body weight will be measured in all subjects at Weeks 26 and 78 as well. Milestone data will include weight and BMI; body composition (BIA, skinfolds) and fat distribution (WHR); blood pressure; fasting insulin, glucose, and lipid profile; exercise levels; dietary intake (FHFFQ); compliance; psychologic assessment; and intercurrent health events, with particular emphasis on screening for potential gallbladder and other GI problems. The statistical design will permit the study of differences in weight (and other milestone data) among the three groups and of treatment x clinic interactions; specific covariate models for data at each time point and overall (including time/trend covariate models for each treatment group) will also be examined. Specific approaches will include the use of linear models, analysis of covariance, random coefficient regression models, and diagnostics for model fit, for outcome correlation, for modelling the variance, and for influence. We hypothesize that African American women randomly assigned to a culturally sensitive weight-control intervention will reduce their body weight by 5-10% as compared to individuals randomly assigned to a control group and that this weight loss will be sustained for a minimum of one year. Secondary objectives of this study will include monitoring changes in risk factors for disease and comparing two dietary treatment interventions (MRD vs FBD) for compliance and acceptability; based on our prior experience, we hypothesize that the MRD will produce better weight loss and maintenance than the FBD. In addition, this research should develop a turnkey program that can be administered through church and community centers by trained African American lay persons.